Clinical question: Do preoperative hemoglobin A1c (HbA1c) and perioperative glucose predict outcomes in patients undergoing noncardiac and cardiac surgeries?
Background: Hyperglycemia in the perioperative period has been associated with infection, delayed wound healing, and postoperative mortality. Studies have investigated the effects of HbA1c or hyperglycemia on postoperative outcomes, but none have been performed to assess the effect of one while controlling for the other.
Study design: Retrospective analysis.
Setting: Single-center, Duke University Health System.
Dr. Antony Agith
Synopsis: Using a database of electronic health records at Duke University Health System, Durham, N.C., investigators reviewed 13,077 surgeries (6,684 noncardiac and 6,393 cardiac) to determine the association of preoperative HbA1c with perioperative glucose and 30-day mortality. For noncardiac surgery, increased average perioperative glucose was associated with increased mortality (P = .04). In cardiac surgery both low and high average glucose was associated with increased mortality (P = .001). By contrast, HbA1c was not a significant predictor of postoperative mortality in cardiac surgery (P = .08), and in noncardiac surgery, HbA1C was negatively associated with 30-day mortality (P = .01). Overall, perioperative glucose was predictive of 30-day mortality, but HbA1c was not associated with 30-day mortality after researchers controlled for glucose.
Because the study is retrospective, no causal relationship can be established. Hospitalists involved in perioperative care should aim for optimization of glucose control regardless of preoperative HbA1c.
Bottom line: Perioperative glucose is related to surgical outcomes, but HbA1c is a less useful indicator of 30-day postoperative mortality.
Citation: Van den Boom W et al. Effect of A1C and glucose on postoperative mortality in noncardiac and cardiac surgeries. Diabetes Care. 2018 Feb;41:782-8.
